# NousResearch/hermes-agent · Cache token fields missing from API server usage events

## Open now

- 2 issues, 3 PRs open on GitHub.
- #25400: api_server.py strips cached_tokens from OpenAI-compat usage emission
- #29553: Bug: cache_read_tokens and cache_write_tokens missing from API Server SSE usage events
- PR #14823: feat(api_server): surface prompt-cache tokens and model in usage block
- PR #25428: fix(api_server): preserve cached_tokens in OpenAI-compat usage emission
- PR #29554: fix: include cache_read_tokens and cache_write_tokens in API Server SSE usage events

A bug was identified where cache_read_tokens and cache_write_tokens were absent from API server SSE usage events and related usage reporting paths. Multiple fixes address including these fields in SSE events, OpenAI-compatible usage emission, and agent run usage tracking.

## TL;DR

- 7 member(s): 2 open issue(s), 0 closed issue(s), 3 open PR(s), 0 merged PR(s), 2 PR(s) closed without merging.
- No fix has been merged to main; one or more PRs were attempted (see below).
- **Recommended:** review the prior/competing PR(s) below. A fix was attempted but not merged; your case may extend or revive that work.

## Maintainer resolution verdict

- Internal-tracker verdict over this cluster's issue members: `unresolved`. This is distinct from the fix status above. Weigh it yourself.

## How this cluster was grouped

- Link evidence (`STRUCTURAL_PARTIAL`): some members are connected by structural evidence (shared changed files, or a closing/superseded reference); the rest are joined by an automated relevance judgement over their text.
- Members outside that structural core may belong to a different underlying problem.
- Pairwise link density `0.26`: the share of member pairs that carry a link, counting a structural link in full and a relevance-only link at 0.36. `1.00` means every pair is linked.

## Open related issues

- #25400: api_server.py strips cached_tokens from OpenAI-compat usage emission
- #29553: Bug: cache_read_tokens and cache_write_tokens missing from API Server SSE usage events

## Prior & competing PRs

- **#25472: fix: surface cached_tokens in OpenAI-compat usage responses (#25400)**. closed without merge · fix tier `suggested`
  - maintainer direction: category `uncategorized`; implication `needs_llm`; refs `#25422`, `#25428`, `#25400`, `#14823`
  - _suggested-tier fix detail withheld; re-request with `?tier=all` to include it._
- **#14823: feat(api_server): surface prompt-cache tokens and model in usage block**. open
  - maintainer direction: category `uncategorized`; implication `needs_llm`; refs `2710ee550433`, `2710ee550433`
  - files: `gateway/platforms/api_server.py`, `tests/gateway/test_api_server.py`, `website/docs/user-guide/features/api-server.md`
- **#25422: fix(api_server): surface cached_tokens in OpenAI-compat usage emission**. closed without merge
  - maintainer direction: category `implemented_elsewhere`; implication `superseded`
- **#25428: fix(api_server): preserve cached_tokens in OpenAI-compat usage emission**. open
  - maintainer direction: category `duplicate_of`; implication `duplicate`; refs `#25428`, `#25422`, `#14823`
  - files: `gateway/platforms/api_server.py`, `tests/gateway/test_api_server_usage.py`
- **#29554: fix: include cache_read_tokens and cache_write_tokens in API Server SSE usage events**. open
  - maintainer direction: category `duplicate_of`; implication `duplicate`; refs `#25422`, `#14823`, `#25673`, `#25717`, `#25891`, `#23665`, `#25428`, `#29553`
  - files: `gateway/platforms/api_server.py`

## Fix status

- No merged fix on record.
- Subsystem: `gateway`